Comments (11)
Hi Yuling,
At this time, we're still working on making MANA really robust. Our primary work environment for now is Cori at NERSC. For CentOS 7, we've partially ported it so that it can launch and checkpoint, but it cannot yet restart. As the project becomes more mature on Cori, we will roll it out to other platforms.
The safest thing is to check back later.
You wrote that you're using intel open mpi. Is this Intel MPI or Open MPI? For certain technical reasons, we require the ability to create a statically linked MPI executable. We will fix this later. From what I understand, Open MPI does not support statically linked executables (i.e., does not support libmpi.a).
from mana.
@gc00 Hi gc00,
The kernel of the HPC in my school is CentOS 7, and the MPI version is Intel MPI. For now, I try to checkpoint the MPI job on HPC. By using DMTCP, I was able to checkpoint/restart single-node MPI jobs. However, for multi-node MPI jobs, as you said, DMTCP can only save checkpoints, but cannot restart.
When will you be able to support Linux CentOS 7? Thank you
from mana.
I haven't tried MANA yet because I had some problems installing it on CentOS 7.
/usr/bin/ld: cannot find -llzma
/usr/bin/ld: cannot find -lxml2
I don't have root permission of our HPC, so I can't install the missing library. I'm working on trying to use MANA on HPC.
from mana.
Hi Yulin,
The MANA for CentOS 7 is still a work in progress. But it will be possible to download the rpm packages and extract what you need there for the missing libraries.
I estimate another two weeks or so before I have it fully ported.
Best,
- Gene
from mana.
Hi Gene.
I will try to download the rpm package for the missing libraries. Thank you very much for your help.
And I am looking forward for the fully porting on CentOS 7.
Yulin
from mana.
Hi Gene.
I will try to download the rpm package for the missing libraries. Thank you very much for your help.
And I am looking forward for the fully porting on CentOS 7.Yulin
Hi,
I had the same problem, did you solve it?
from mana.
Hi Yulin,
The MANA for CentOS 7 is still a work in progress. But it will be possible to download the rpm packages and extract what you need there for the missing libraries.
I estimate another two weeks or so before I have it fully ported.
Best,
- Gene
Hi, Gene
I have tried to use openMPI, Intel MPI, and MPICH to compile mana, but have encountered many problems. Which MPI should I use to compile mana, or is it because my environment is CentOS7?
from mana.
Hi Gene.
I will try to download the rpm package for the missing libraries. Thank you very much for your help.
And I am looking forward for the fully porting on CentOS 7.
YulinHi,
I had the same problem, did you solve it?
I solved this on the direction of a file whose location is mana/contrib/mpi-proxy-split/lower-half/README.mpich-static. After solving this, I have other questions about installing mana on CentOS 7. And i don't fix it yet.
from mana.
@gc00 Hi Gene,
have you successfully port MANA to Centos 7?
Also, for multi-node MPI jobs, DMTCP cannot restart them. I hope that DMTCP can support multi-node MPI checkpoints.
Best,
Yulin
from mana.
@gc00 Hi gc00, The kernel of the HPC in my school is CentOS 7, and the MPI version is Intel MPI. For now, I try to checkpoint the MPI job on HPC. By using DMTCP, I was able to checkpoint/restart single-node MPI jobs. However, for multi-node MPI jobs, as you said, DMTCP can only save checkpoints, but cannot restart. When will you be able to support Linux CentOS 7? Thank you
HI? how about your process? I compile the MANA succeed with MPICH in CentOS7,and save checkpoints with hellpmpi.c in local node,but when restart it fails.
from mana.
@gc00 How is the status on OpenMPI supprt in MANA? Thanks!
from mana.
Related Issues (20)
- Error installing mana on SuSE cluster
- Wrapper for MPI_Type_hvector needed for iPic3D HOT 1
- Problem with the MPI_Type_hvector wrapper for iPic3D HOT 5
- Handling completed asynchronous request mappings HOT 2
- Implemented wrapper for MPI_Cart_create() not working with iPic3D HOT 3
- MANA build fails on CentOS 8 HOT 3
- Final issues with iPic3D HOT 7
- MPI_Scan Implementation is Incorrect for MPI_COLLECTIVE_P2P Defined
- Unable to checkpoint: Checkpointing during dense collective calls hangs HOT 7
- Cannot build Mana on container CentOS 7.9 HOT 3
- Are messages on the fly lost after checkpointing?
- Singularity / apptainer / container support
- Latest commit does not compile. make fails with "no known conversion from 'const void*' to 'MPI_Datatype' {aka 'ompi_datatype_t*'}" HOT 4
- mana_restart segfaults if no ckpt dir present
- `make -j mana` can fail with `p2p_deterministic.c`, where `make mana` does not. HOT 3
- `./configure-mana` enables debug by default, where just running `./configure` apparently does not. HOT 3
- `mpi-proxy-split/util/mpi-logger` will not work for OpenMPI.
- `./util/readdmtcp.sh` will print out a fake `vvar` even on systems where it does not exist
- License
- CUDA support
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mana.